Language attitudes in fast-growing societies: new insights in the dynamism dimension

The study investigates whether attitudinal dimensions commonly reported in language attitude research (e.g., status vs. solidarity; status, solidarity and dynamism) hold in fast-growing, rapidly urbanising societies, using Saudi Arabia as a case. Against the backdrop of Saudi Arabia’s significant social and cultural reforms under Vision 2030 and diverse regional Arabic dialects, the research aims to uncover: (1) the attitudinal dimensions underpinning Saudi language attitudes and how dialects associate with these dimensions; and (2) the effects of socio-demographic variables (dialect, gender, age, education, place of birth) on these dimensions. By applying an indirect attitudinal elicitation method, the study seeks to clarify how cognition-driven associations map dialects to evaluative traits in a context of rapid social change, and to assess implications for education and media in reducing dialect prejudice.

The paper situates language attitudes (LAs) within a tripartite structure of cognition, affect, and behavior, emphasizing cognition in uncovering latent evaluative dimensions that link dialects to characteristics. It distinguishes LA research from language ideology work, while noting their interaction. Methodologically, indirect techniques such as the matched guise (MGT) and verbal guise (VGT) uncover covert attitudes; the study adopts VGT to avoid demand characteristics and because accurate dialect mimicry is difficult in MGT. To ensure participants evaluate the intended variety, a dialect recognition task, as advocated by Preston, is included. Classical LA findings often reveal two dimensions—status (prestige/correctness) and solidarity (social attractiveness)—with standard varieties typically higher on status and in-group loyalty linked to solidarity. A third dimension, dynamism (e.g., energetic/enthusiastic), has been identified (Zahn & Hopper, Kristiansen), and may gain prominence in contexts of linguistic and social change. In Saudi Arabia, prior work has been limited in scope: small-scale or region-specific attitude studies (e.g., Alrumaih 2002; Aldosaree 2016), sub-dialect studies (Meccan, Qasimi), and preliminary national keyword studies (Alhazmi & Alfalig 2022). Alhazmi (2018) found a modern–traditional dichotomy in Hijaz due to its distinctive sociolinguistic history. Broader perceptual dialectology studies map dialect areas and stereotypical labels (Al-Rojaie 2021). Saudi dialects are geographically bounded (Najdi, Northern, Eastern, Western/Hijazi, Southern) within an Arabic diglossic context (MSA vs. dialect). Media portrayals appear to shape stereotypes (e.g., southerners as naïve, Najdis as shrewd). This literature motivates a national, dimensional analysis that accounts for social change and demographic effects.

Design: Online questionnaire with two parts: (1) VGT-based ratings of four male speakers (Najdi, Eastern, Northern, Southern) and a dialect-recognition task; and (2) demographics (dialect, gender, age, education, place of birth). Voice Stimuli: For each dialect, three male speakers (ages 35–45) recorded 1-minute spontaneous speech describing weather from a map prompt to elicit natural speech. Recordings were noise-reduced (Audacity). An eight-judge panel (from respective regions, ≥10 years residence, tribal affiliation) selected the best representative speaker per region. Attempts to include female speakers failed for Northern and Southern due to cultural constraints; thus, all stimuli were male. Measures: Participants first assigned each voice to one of five regions (recognition/allocation task). They then rated each guise on 14 evaluative characteristics via 5-point semantic differential scales (1 = least to 5 = most). Items adapted from Alhazmi & Alfalig (2022): Bedouin, prestigious, calm, beautiful, uncultured, harsh, arrogant, confident, friendly, open-minded, moralistic, easy to understand, difficult to understand, generous. Negatively worded items were reverse-scored. Procedure and Sampling: Distributed via Twitter and WhatsApp (June–July 2021) using convenience sampling targeting Saudis inside/outside the country. Participants: N=568 Saudis; gender: 32% male, 68% female; ages 18–64 (majority 18–34); education: 29.9% secondary, 53.7% undergraduate, 16.4% postgraduate; self-identified dialects and places of birth spanned all regions. The sample is below the calculated requirement (~602) for 95% CI and 4.15% margin of error. Analysis: Three-stage statistical plan in SPSS. (1) Exploratory factor analysis (EFA) to uncover latent attitudinal dimensions. Suitability checks: KMO and Bartlett’s test; extraction via Kaiser’s criterion and scree plot; oblique rotation (direct oblimin) due to correlated factors. (2) Internal consistency via Cronbach’s alpha; items potentially dropped to improve reliability. (3) ANOVA to test effects of dialect, gender, age, education, and place of birth on factor scores; Tukey HSD post-hoc tests for pairwise comparisons.

Dialect recognition: High correct allocations: Eastern 91.9%, Southern 91.0%, Northern 87.7%, Najdi 83.3%. Factor structure: Initial EFA (KMO=0.806; Bartlett p<0.001) suggested three factors (eigenvalues >1) explaining 51.24% variance (F1 25.56%, F2 13.15%, F3 12.53%). Reliability: Cronbach’s alpha—F1=0.712, F2=0.638, F3=0.661. To improve reliability, deleted ‘Bedouin’ from F2 and ‘arrogant’ from F3, yielding F2=0.707 and F3=0.691. Refined EFA (KMO=0.812; Bartlett p<0.001) confirmed three factors explaining 52.80% variance (F1 29.22%, F2 13.53%, F3 10.05%). Factor interpretation (oblimin rotation): - F1 (Dynamism–Status): open-minded, confident, prestigious, easy to understand, beautiful. - F2 (Dynamism–Language): difficult to understand, harsh, uncultured, easy to understand (language comprehensibility/vitality focus). - F3 (Solidarity): generous, moralistic, calm, friendly. Dialect means by factor (1–5 scale): - Southern: Dynamism–Status 2.69; Dynamism–Language 3.22; Solidarity 3.53. - Northern: 3.44; 3.74; 3.59. - Eastern: 3.35; 3.64; 3.32. - Najdi: 3.70; 3.86; 3.42. Pattern: Najdi rated highest on both dynamism-type dimensions; Southern lowest on dynamism. Northern highest on solidarity; Eastern lowest on solidarity. Demographic effects (selected): - Own dialect: In-group loyalty evident, especially for Southern and Najdi on dynamism–status; Najdi also higher on dynamism–language (ANOVAs p<0.001; Tukey details reported). - Gender: Females rated northern, eastern, Najdi higher than males on dynamism–language; females also more positive on solidarity and on Najdi in dynamism–status (all p<0.001, small-to-moderate effect sizes up to η2≈0.13). - Education: Higher education associated with more positive ratings for northern, eastern, and Najdi on dynamism–language, and for Najdi on dynamism–status; conversely, lower education (secondary) rated Southern higher on dynamism–status; undergraduates rated Southern higher on solidarity than secondary (p<0.001). - Place of birth: In-group loyalty for Southern and Najdi on dynamism–status; broader positive perceptions of northern and eastern (dynamism–language) from those born in Eastern, Najdi, and Hijazi regions; Najdi-born rated Najdi higher on both dynamism dimensions (p<0.001). Overall: A three-dimensional model emerges where dynamism predominates (split into two interrelated facets), challenging the traditional primacy of status in fast-growing societies. Education appears to attenuate bias; women exhibit greater acceptance across dialects.

The findings address RQ1 by revealing a three-dimensional evaluative model with two dynamism-related factors (dynamism–status and dynamism–language) plus solidarity, indicating that in a rapidly urbanising context, dynamism rather than status organizes attitudinal space. This aligns with work showing the salience of dynamism amid linguistic innovation and emerging standards (e.g., Danish neo-standard). The semantics show that prestige and aesthetic appeal (‘beautiful’) are integrated into dynamism–status, while comprehensibility and perceived harshness/unculturedness load onto dynamism–language, reflecting vitality and usability. Associations by dialect show Najdi as the most dynamic (consistent with perceptions of Riyadh/Najdi as locus of national reform and a koine/standard), Southern as least dynamic, and Northern as most solidary with relatively small solidarity differences overall, suggesting maintained social cohesion. RQ2 is addressed by significant demographic effects: (a) robust in-group loyalty, particularly for Southern and Najdi (notably mapped to dynamism rather than only solidarity, extending prior patterns); (b) women’s higher acceptance/positivity toward multiple dialects; and (c) higher education correlating with less biased, more positive evaluations, especially on dynamism dimensions. These findings are theoretically significant: they challenge the presumed dominance of status–solidarity in favor of dynamism–solidarity in fast-growing societies, and suggest that social reforms and emerging standards may elevate dynamism as the primary evaluative lens. Practically, they underscore the potential of educational interventions and revised media portrayals to mitigate dialect prejudice and support linguistic diversity.

The study contributes a novel, three-dimensional attitudinal model from Saudi Arabia, where dynamism predominates over status, bifurcating into dynamism–status and dynamism–language, alongside a solidarity dimension. Najdi is most strongly associated with dynamism; Southern least so; Northern leads in solidarity. Demographic analyses reveal in-group loyalty (especially Southern and Najdi), greater acceptance among women, and education’s role in reducing bias and enhancing positive attitudes. Implications include (1) integrating curricula that highlight the cultural value of dialect diversity to reduce prejudice; and (2) encouraging media to represent dialects based on actual social practices rather than stereotypes. Future research should employ larger, representative samples, include balanced male/female stimuli, and address the complex Hijazi dialect situation with separate Bedouin and Hadari representations to enhance generalizability.

- Voice stimuli were all male due to refusal of some female speakers (Northern and Southern) to be recorded; gender imbalance may affect attitudinal responses, particularly regarding dynamism. - Convenience sample (N=568) below the calculated requirement (~602) for 95% CI and 4.15% margin of error; not representative of the ~35 million Saudi population, limiting generalizability. - Exclusion of the Hijazi dialect due to its internal dichotomy (Bedouin vs. Hadari) and complex sociolinguistic situation; omitting a major region may limit comprehensiveness of national inferences.